rebound
rebound copied to clipboard
Switching integrator after archive reload
Hi Hanno,
Hopefully this is a quick one.
Any reason why switching from mercurius to IAS15 or WHFast after reloading a simulation may cause a segmentation fault or ipython kernel death upon integration? The problem occurs whether I'm setting the new integrator in python or C.
I'm trying to optimize a particular simulation that I'm going to have to run a bunch of times and it would be great to test out different integrators continuing from an initial state that I had previously saved.
Maybe I need to rebuild the simulation "Manually"?
Best, -Robert
Hm. This shouldn't be a problem (and segmentation faults shouldn't happen at all). But I haven't tested it a lot myself. Do you have a short example that reproduces the problem?
Hi, Sorry I haven't moved forward with an example for this problem yet...
It is still active, at least for my bin files, but I have a workaround that is working for now.
I will try to generate an example soon.
I'm closing this for now, but feel free to reopen the issue if a problem still persists.